Intelligent Authoring Platform (IAP)
Welcome
Welcome to the Intelligent Authoring Platform — Roche's enterprise solution for creating, managing, and publishing structured documents. IAP brings together intelligent authoring, automated data integration, and AI-powered content generation in a single platform, enabling teams to produce high-quality, compliant documents faster and more consistently.
What is IAP?
The Intelligent Authoring Platform is a web-based application where teams collaboratively author, review, and release structured documents. Unlike traditional word processing tools where authors manage both content and formatting, IAP separates what you write from how it looks — allowing you to focus on meaningful content while the system handles formatting, consistency, and output generation automatically.
Built on structured content
At its core, IAP is a structured content authoring system. Structured content is information that has been organised in a predictable, standardised way so that each piece of content is defined by its purpose and meaning — not just its visual appearance. Unlike traditional documents where text is a flat stream of characters, structured content uses metadata, tags, and predefined schemas to describe what each element is (e.g., a section title, a product name, a regulatory reference) rather than how it looks.
In IAP, everything you work with — templates, documents, components, sections, smart fields, and metadata — is structured content. This is what enables the platform to automate formatting, enforce consistency, reuse content across documents, and generate outputs in multiple formats from a single source.
The role of XML
All content in IAP is stored as XML (Extensible Markup Language) — a machine-readable, open-standard format where every piece of information is tagged by its function. When you type a component title, insert a smart field, or add a table, you are building a structured, semantic data model that the system understands and can process intelligently.
| Benefit of XML | What it means for you |
|---|---|
| Separation of content and presentation | You focus on writing. The system applies formatting automatically at output time based on the template style you select. |
| Multiple outputs from one source | The same content can be rendered as DOCX, PDF, or XML — each with different styles — without rewriting anything. |
| Machine-readable content | The system can automate tasks: validating completeness, pulling smart field values, filtering by conditions, and generating tables of contents. |
| Long-term preservation | XML is text-based and system-independent. Content remains accessible regardless of which tools or platforms are used in the future. |
| Reusability | Structured components can be shared and linked across templates and documents — because the system knows exactly what each piece of content represents. |
You don't need to know XML. The editor presents a visual, user-friendly interface — you write and edit content just like in any modern authoring tool. The XML structure is created and maintained automatically behind the scenes.
The shift from traditional authoring
| Aspect | Traditional documents | Structured content in IAP |
|---|---|---|
| Author's focus | Both information and visual presentation | Meaning only — formatting is automatic |
| Storage format | Proprietary binary files (.docx) | Open-standard XML — semantic and future-proof |
| Consistency | Manually checked; prone to drift | System-enforced through templates, metadata inheritance, and smart fields |
| Content updates | Manual copy-pasting across multiple files | Single source of truth — update once, propagate everywhere via reuse |
| Searchability | Basic keyword matching | Metadata and smart fields make content queryable by specific data types |
| Compliance | High risk of human error and missing sections | Predefined templates with mandatory components and conditions highlight gaps |
| Formatting | Time-consuming manual styling | Automated — select a template style and the system handles the rest |
| Output formats | Manual conversion for each format | One-click generation of DOCX, PDF, and XML |
| Data integration | Manual copy-paste from external sources | Smart fields and smart tables automate data import with built-in filtering |
| System transfer | Manual upload to downstream repositories | Integrated transfer to Veeva QualityDocs, Veeva RIM, and eVAL Roche |
| Collaboration | Sequential; one person edits at a time | Component-level review — multiple reviewers work in parallel |
How IAP features apply structured content principles
| IAP feature | Structured content principle |
|---|---|
| Templates | Define the content architecture — which sections exist and what structure documents must follow. |
| Components | Break content into independently managed, reusable building blocks — each with its own status, version, and lifecycle. |
| Sections | Provide sub-structure within components, organising content by meaning rather than visual layout. |
| Metadata (namespace, workspace, template, document) | Tag content with structured, queryable information — inherited across levels to ensure consistency. |
| Smart fields | Pull data dynamically based on metadata values, eliminating manual copy-pasting and ensuring accuracy. |
| Conditions | Control which content is included in the output based on metadata values — enabling one template to serve multiple use-cases. |
| Component reuse | Enable a single source of truth — author content once and share it across templates and documents with live links. |
| Preview & Output | Separate content from presentation — the same structured content can be rendered as DOCX, PDF, or XML with different template styles, all automatically. |
Layered content architecture
IAP is built around a layered architecture where each level inherits configuration and metadata from the one above:
| Layer | Purpose | Example |
|---|---|---|
| Namespace | Separates use-cases and environments | |
| Workspace | Groups related documents under a project or product | |
| Template | Defines the structure and rules for a document type | |
| Document | The actual working output — authored, reviewed, and released |
Each layer inherits configuration and metadata from the one above, creating a cascade of consistency from namespace down to individual documents.
How this guide is organised
| User Guide Section | What you'll learn |
|---|---|
| Namespaces | How namespace environments work and how to switch between them. |
| Workspaces | How to create and manage project containers/folders, assign members, and configure metadata. |
| Templates | How to create and configure templates — including metadata and GenAI setup. |
| Documents | How to create documents from templates, fill in metadata, select components for reuse from other documents. |
| IAP Editor | All editor features: components, sections, text formatting, tables, images, equations, smart fields, conditions, notes, links, citations, footnotes, comments, tasks, author guidance, GenAI Assistant, etc. |
| Review & Release | How to send components for review, collaborate through comments and tasks, resolve feedback, and release templates and documents. |
| Preview, Output & Transfer | How to generate previews, save outputs, and transfer documents to external systems. |
Getting started
Ready to begin? Here's a quick path based on your role:
If you are an Author:
- Select your namespace
- Create or join a workspace
- Create a template or Create a document
- Start authoring in the editor
If you are a Reviewer:
- Select your namespace
- Open a template or document assigned to you
- Review content and leave comments
Need help? If you have any issues oor questions not covered by this guide, reach out to your IAP support team via RoSE portal